some bad cases about co-mod-gan HOT 5 OPEN

Modeling long-tail distributions is generally hard for generative models. With the current state of the art, I'm afraid that the only solution is to reduce the variety of the dataset to make it work better on a specific type of images.

from co-mod-gan.

Hi @yezhanglang can you share how did you manage training?

from co-mod-gan.

@zsyzzsoft
I'm facing worse result after 2960k iter. Any ideas？
https://s2.loli.net/2022/03/13/u8qtRLrPWwgBGyZ.png
Besides, is the fake_init normal? Seems the masks are not binary.
https://s2.loli.net/2022/03/13/O7SwvqP9Mh3TmKl.png

from co-mod-gan.

If the FID metric suddenly spikes up at some iteration, you can try resume training from the last normal checkpoint. Sometimes this happens because of GPU issue. The fake_init looks normal.

from co-mod-gan.

If the FID metric suddenly spikes up at some iteration, you can try resume training from the last normal checkpoint. Sometimes this happens because of GPU issue. The fake_init looks normal.

Thanks for your quick reply! The FID smoothly drops. But all the training samples seem bad, I have no idea what happened as the fake_init looks normal.

from co-mod-gan.